Catalytic CO Oxidation over Au Nanoparticles Supported on Yttria ...
ECS Transactions, 45 (1) 265-274 (2012) 10.1149/1.3701316 © The Electrochemical Society
Catalytic CO Oxidation over Au Nanoparticles Supported on Yttria-Stabilized Zirconia Holly A.E. Dolea, Jong Min Kima, Leonardo Lizarragab, Philippe Vernouxb, Elena A. Baranovaa* a
Department of Chemical and Biological Engineering, University of Ottawa, 161 LouisPasteur St. Ottawa, ON, K1N 6N5 Canada b Université Lyon 1, CNRS, UMR 5256, IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon, 2 avenue Albert Einstein, F-69626 Villeurbanne, France *
email : [email protected]
Gold nanoparticles supported on yttria-stabilized zirconia (YSZ) were synthesized using a modified alcohol method which involves the stabilizing agent, polyvinylpyrrolidone (PVP). Two different average sizes (13.1 and 17.1 nm) of Au nanoparticles were synthesized. These catalysts were tested for gas phase CO oxidation. Since PVP is known to block the active Au sites, the catalysts were calcined at 300°C and 600°C in order to remove the PVP. Overall, higher catalytic activity was found for the smaller Au nanoparticles. It was also found that calcination is required in order to achieve activity, even though the particle size increases with this treatment. The effect of calcination temperature did not prove to be significant. It is demonstrated that O2- ionically conductive YSZ is a promising catalyst support that can finely disperse and stabilize Au nanoparticles for CO oxidation. Introduction For many years, gold has been disregarded as an active catalyst in its bulk state due to its chemically inert property. It wasn’t until Haruta (1-3) discovered that when gold particles with diameters smaller than 10 nm are supported on metal oxides, they tend to be very catalytic active for many reactions. However, this discovery came with the challenge of achieving small Au nanoparticles as Haruta stated, due to quantum size-effects, the smaller the Au nanoparticle, the lower the boiling point which results in significant agglomeration during calcination (1). Recently, research has been focused on determining the effect of size and pretreatment conditions (i.e., calcination temperature). It has been found that Au nanoparticles less than 5 nm show a significant increase in catalytic activity (4, 5). Others have also shown that calcination causes agglomeration and that as the temperature is increased the gold phase changes from gold oxide to metallic gold which shows to influence the catalytic activity (6, 7). The effect of the support material has been shown to have less of an influence on the activity (5) which could allow for flexibility depending on the desired application. The Au nanoparticles used in this study were supported on yttria-stabilized zirconia (YSZ), a metal oxide support. YSZ is known to be an O2-conductor due to the presence of oxygen vacancies inside its crystallographic structure. This oxide has been extensively studied as a material for oxygen sensors and fuel cells (8). Recently, YSZ has received attention as a catalytic support especially in studies of Electrochemical Promotion of Catalysis (EPOC); also called the NEMCA (Non Faradaic Electrochemical Modification of the Catalytic Activity) effect (9). Vayenas et al. (9) were the first to show that the migration of ionic species from a solid
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS265 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
electrolyte to the catalyst surface induced by electrical polarizations can improve catalytic performances. Vayenas et al. (10) have described EPOC as an electrically controlled strong metal support interaction (SMSI). Recently it was shown that YSZ is an attractive catalyst support for nanoparticle stabilization and activation for the reaction of CO oxidation over Pt/YSZ nanoparticles (11-13). It was found that O2- ionic species from YSZ can significantly promote the catalytic activity of Pt nanoparticles due to the thermally induced O2- backspillover from YSZ over Pt catalyst in agreement with the electrochemical promotion mechanism. It was also found that observed effect becomes more pronounced with decreasing the particle size (12). In this work, gold nanoparticles supported on YSZ (Au/YSZ) are investigated for the first time in the reaction of CO oxidation. To this end, Au nanoparticles with two different average sizes (13.1 and 17.1 nm) were synthesized using a modified alcohol method. This method was modified to achieve polyvinylpyrrolidone- (PVP) stabilized Au nanoparticles supported on YSZ. The variation in nanoparticle size was achieved by adjusting the ratio of Au to PVP in the synthesis solution (14-16). The particle sizes of these catalysts were studied using x-ray diffraction (XRD). Surface properties were also characterized using transmission electron microscopy (TEM). An effective calcination temperature was also determined for the removal of PVP using a thermogravimetric analysis (TGA). The PVP-stabilized Au/YSZ nanoparticles were tested using the model reaction of CO oxidation. The effect of the Au particle size and calcination temperature on the catalytic activity was investigated. Experimental Synthesis of Au Nanoparticles Gold nanoparticles (Au), deposited on YSZ powder (Au/YSZ) were synthesized using a similar method to the colloidal methods using alcohol as a reducing agent (17-19); however, the procedure was modified to achieve supported Au/YSZ. Gold (III) chloride (Aldrich, 99.999%) precursor salt was dissolved in a mixture of nanopure water (18 MΩ cm) and ethanol (Fischer, 99%) with a ratio of water to ethanol of 1:1, in the presence of PVP (Sigma, 10,000 average molecular weight) and YSZ powder (TOSOH, 8% YSZ, with average crystallite size of 25 nm, 0.6 µm particle size and bulk density of 1.3 g·cm-3). The molar ratio of PVP to gold was 3:1 (Au1) and 10:1 (Au2) for the samples tested. The desired gold loading was 1 wt. % on YSZ. The obtained mixture was stirred for 30 minutes at room temperature then refluxed at 100°C for 2 h, followed by extensive washing with nanopure (18 MΩ·cm) water and drying in the air at 70°C. The initial color of the cloudy mixture appears yellow. After about 30 minutes, the colour of the mixture turns pink or purple, depending on the concentration of PVP. Upon drying, the synthesized 1wt. % Au nanocatalysts on YSZ powder have a pink to purple color. PVP is used as a polymer protective agent in order to achieve small nanoparticles by stabilizing the Au ions during the synthesis (14-16). It is believed that the Au nanoparticles are stabilized through the reduction of van der Waals interaction between colliding colloidal particles (14). Before testing the performance of the catalysts, the PVP must be removed as it known to inhibit the catalytic activity (20, 21). This was done by calcining the Au/YSZ catalyst at 300 or 600°C for 2 h in air.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS266 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
Transmission Electron Microscopy Transmission electron microscopy (TEM) images were taken for the Au/YSZ catalysts tested (JEOL 2010 LaB6) in order to obtain a higher resolution so that the morphology and size of Au particles may be better investigated. An extraction replica technique was used for sample preparation (22). The catalyst was dispersed in ethanol, deposited on a mica film and covered with a carbon layer. The YSZ support and film of mica were then dissolved in hydrogen fluoride solution for 24 hours resulting in Au particles being fixed on the carbon film. These particles were directly observed by TEM. The use of Image J software allowed for the determination of Au particle size distribution. X-ray Diffraction X-ray diffraction (XRD) patterns of the Au/YSZ nanoparticles were collected using RigakuUltima IV diffractometer using Cu Ksource. The experiments were run in the focused beam geometry with a divergence slit of 2/3 degree, a scan speed of 0.17 degree·min-1 and a scan step of 0.06 degree for all experiments. The diffractograms were collected between 30 and 55°2. The average particle diameter was determined using the Scherrer equation and the full widths at half maximum (FWHM) of the (111) plane of gold at 38° 2. Thermogravimetric Analysis Thermogravimetric analysis (TGA) (2960 SDT V3.0F) was carried out to determine an optimum temperature range at which the synthesized Au/YSZ samples are to be calcined in order to completely remove the PVP stabilizing agent. The temperature range should be high enough to decompose the carbons and other volatile compounds in the PVP polymer, but also, the temperature should be low enough that sintering and formation of agglomerates of Au nanoparticles are minimized. In order to determine the proper calcination temperature, TGA is run to establish the temperature range at which carbon decomposes. The sample is heated at a rate of 1.2°C min-1 from room temperature to 580°C. Catalytic Activity Measurements The catalytic activity measurements of the Au/YSZ catalysts were carried out at atmospheric pressure in a continuous flow quartz reactor. The reaction gases were CO (Linde, 1.02% CO in He) combined with pure O2 (Linde, pure O2), and pure He (Linde, pure He) as a carrier gas. The gas compositions were controlled by mass flow controllers (Brooks, 5850 TR Series and MKS electronic mass flow controllers) and were the following: 760-780 ppm CO with 4.3-4.5% O2 and He balance for CO oxidation. The overall flow rate was held constant at 4.3 L·h-1, resulting in a space velocity of approximately 19 000 h-1. The product gases for the oxidation of CO were analyzed with an on-line micro-chromatograph (μGC-R3000 SRA instruments) which contained two separate columns; one to separate O, N2 and CO at a temperature of 90°C and the other to separate CO2 at a temperature of 100°C. CO2 concentration in the product gas of the reactor was also monitored by an infrared (IR) analyzer (Horiba, VA3000). A K-type thermocouple was attached to the reactor at the catalyst bed in order to monitor the temperature during the heat-up period. The entire reactor was placed in a furnace, which was attached to a temperature controller. The sample was heated from 25 to 250°C at a rate of
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS267 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
3°C/min. For each catalyst, three heating cycles were performed to observe the stability of the catalyst. Results and Discussion Characterization of Au Catalysts Figure 1 shows TEM images of Au2 which was untreated and calcined at 600°C. This image shows that the Au nanoparticles are uniformly distributed over the YSZ support. From the corresponding histograms, an average particle diameter was found to be 9.1 and 11.7 nm, respectively. These results show that calcination does have an effect on the particle size.
(a)
Percentage of Particles (%)
18 15 12 9 6 3 0 3
4
5
6
7
8
9 10 11 12 13 14
Particle Diameter (nm)
(b)
Percentage of Particles (%)
18 15 12 9 6 3 0 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Particle Diameter (nm)
Figure 1. TEM images (left) and corresponding histograms (right) of Au2 (a) untreated; and (b) calcined at 600°C.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS268 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
The XRD spectrum for each Au/YSZ catalysts is shown in Figure 2. The crystallite size for each sample was determined from the Au (111) peak using the Scherrer equation. These values are summarized in Table 1. It is evident that the size of the crystallite increased (approximately double) with calcination. Similar trends have been found in past studies (6, 7) where it is shown that calcination causes the gold nanoparticles to agglomerate as the melting temperature of the gold decreases with decreasing particle size.
Relative Intensity
4 3
(a) (b) (c) (d)
2 1 0 37 38 39 40 41 2[theta] (deg)
Figure 2. XRD spectra of Au (111) peak of Au/YSZ catalysts (a) Au2 calcined at 600°C; (b) Au1 calcined at 600°C; (c) Au2 untreated; and (d) Au1 untreated. Dispersion calculations were performed for both Au/YSZ catalysts using the average particle diameter from the XRD measurements and the following equation (23), dispersion (%) (MWAu 600) ( Au davg aAu Na )
[1]
where MWAu is the molecular weight of Au (196.97 g·mol-1), ρAu is the density of Au (19.32 g·cm-3), davg is the average particle size, aAu is the surface area of one Au atom (2.29 x 10-19 m2·atom-1), and Na is Avogadro’s number (6.022 x 1023 atom·mol-1). It is evident that since the crystallite size changes between untreated and calcined catalysts (approximately doubled), the dispersion also changes. However, these differences in dispersion do not appear to influence the catalytic activity as Au2 has lower dispersion but higher activity than Au1 (Figure 3). Table 1 summarizes these dispersion values as well as the particle sizes for each catalyst.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS269 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
1st Cycle 2nd Cycle 3rd Cycle Blank Experiment
CO conversion (%)
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 3. Catalytic performance for CO oxidation of Au2, calcined at 600°C. TABLE I. Summary of Au/YSZ Catalysts Prepared and Corresponding Characterization Data. CO Oxidation Avg. Particle Turnover Dispersion Activation Diameter Frequency, TOF Catalyst Pre-treatment (%) Energy, (s-1) (x10-3) davg (nm)* Ea (kJ·mol-1) 100°C 200°C Au1 untreated 8.4 7.1 0 0 77.0 Au1 calcined at 300°C 56.1 Au1 calcined at 600°C 17.1 3.4 0 78.7 63.6 Au2 untreated 6.3 5.3 0 0.25 93.2 Au2 calcined at 600°C 13.1 2.6 8.45 181.74 24.1 *from XRD measurements The result from the TGA of Au1 is shown in Figure 4. The first mass loss of 11.9 % is related to water evaporation and the second mass loss of 1.4 % is related to polymer decomposition. There is a slight loss of mass just below 600°C, likely being a partial removal of oxygen from YSZ. In order to evaluate the effect of all these losses detected from TGA, 300°C (removal of PVP only) and 600°C (removal of all species) selected to be the suitable temperature for calcination.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS270 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
19.5 19.0
Mass (mg)
18.5 11.9% 18.0 17.5 17.0 1.4%
16.5 0
100 200 300 400 500 600 o Temperature ( C)
Figure 4. Thermogravimetric analysis of Au1. Catalytic Performance for CO Oxidation To study the catalytic performance of the Au/YSZ catalysts, the oxidation of CO was used as a model reaction. This reaction has been widely studied due to its environmental and industrial significance (24-27). For each catalyst, the catalytic oxidation of CO was repeated for three successive heating ramps between 25 and 250°C. Figure 3 shows the catalytic activity of Au2 calcined at 600°C. The results showed that the catalyst exhibits stable performance after the first cycle where the activity is observed from around 65°C. It is observed that the first cycle shows lower catalytic activity compared to the successive cycles. Similar multi-cycle performance has been observed by other research groups (28-30); however, the opposite performance has also been observed (31, 32). This observation was explained by the dependence of the pretreatment conditions. In a previous study, it was determined that when catalysts were pre-oxidized, the catalytic activity of the first heating cycle was lower (31). It was explained that when the catalyst is exposed to a reducing species such as CO, the catalyst is activated which is shown by the higher catalytic activity in the successive runs. Another explanation could be the result of residual PVP even after calcination. It is possible that during the first cycle, there is PVP inhibiting the reaction, and it is only after the first cycle that the remaining PVP is removed. Effect of Particle Size Figure 5 shows the effect of Au particle size on the catalytic performance for CO oxidation. Two different ratios of PVP:Au were used to control the size of the particles which has also been shown in previous studies (16, 21). It was explained that the PVP can slow the diffusion of the gold ions within the solution, preventing them from coagulating; therefore, more PVP creates less diffusion. However, it has been shown that there is a limit to the amount of PVP that can influence the particle size (16). From this study, it is clear that the smaller gold
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS271 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
nanoparticles (Au2) synthesized using a ratio of 10:1 (PVP:Au) have the highest catalytic activity.
CO Conversion (%)
17.1 nm 13.1 nm Blank Experiment
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 5. Comparison of catalytic performance for CO oxidation of Au/YSZ catalysts, calcined at 600°C with two different average particle size. Effect of Calcination Temperature Since PVP is known to inhibit catalytic reactions (20, 21), the prepared Au/YSZ catalysts were pretreated by calcining at elevated temperatures to remove this stabilizing polymer. However, calcination can also have a negative effect on the performance of the catalysts. It has been previously shown that higher calcinations temperatures cause agglomeration of the nanoparticles (6, 7). Since smaller nanoparticles are desired for greater activity, the amount of PVP and calcination temperature must be optimized to achieve the best performance. Figure 6 shows a comparison of Au1 that was untreated, calcined at 300°C and calcined at 600°C with respect to a blank experiment (i.e. reactor without any catalyst). From the figure, it is evident that there is a significant effect on the catalytic activity between the untreated and calcined samples; however, the difference between calcinations temperatures does not appear to have such a significant effect. These results show that PVP does indeed inhibit the reaction. Since the PVP is used to stabilize the Au nanoparticles, it is said to be surrounding the nanoparticles. This in turn prevents the surface of the Au nanoparticles to be exposed resulting in lower catalytic activity. Therefore, it is important to sufficiently remove the PVP without causing significant agglomeration of the nanoparticles.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS272 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
(a) (b) (c) (d)
CO conversion (%)
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 6. Effect of calcination temperature on the catalytic activity for CO oxidation of Au1 (a) untreated; (b) calcined at 300°C; (c) calcined at 600°C; and compared to a (d) blank experiment. Conclusion It is demonstrated that YSZ is a promising catalyst support for Au nanoparticles for the reaction of CO oxidation. Au nano-catalysts of two average sizes were synthesized using a modified alcohol method with a PVP stabilizer. The size of the particles depended on the amount of PVP used – more PVP made smaller particles. The catalytic performance of the untreated and calcined catalysts was evaluated for the model reaction of CO oxidation. It was found that calcination had an effect by increasing the particle size by approximately double; however, this treatment was necessary to remove the PVP stabilizer which is known to inhibit the reaction. This was evident by the better performance of the calcined catalyst compared to the untreated catalyst. The most important factor proved to be the size of the particle, where the smaller nanoparticle (Au2) showed the highest catalytic activity for CO oxidation. This catalyst showed catalytic activity beginning around 90°C. In order to improve this catalytic activity, the size of the Au nanoparticles will need to be decreased in further studies. Acknowledgements The authors would like to thank the Natural Science and Engineering Research Council (NSERC) for financial support, the Centre for Catalysis Research and Innovation (University of Ottawa) for XRD and TGA analysis, and IRCELYON for microscopy services.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS273 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
References [1] [2] [3] [4] [5] [6] [7] [8] [9]
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
[23] [24] [25] [26] [27] [28] [29] [30] [31] [32]
H. Haruta, Surf. Sci., 6(2), 102 (2002). M. Haruta, Chem. Rec., 3(2), 75 (2003). M. Haruta, ChemInform., 35(44), 27 (2004). J.D. Grunwaldt, C. Kiener, C. Wögerbauer and A. Baiker, J. Catal., 181(2), 223 (1999). N. Lopez, J. Catal., 223(1), 232 (2004). E.D. Park and J.S. Lee, J. Catal., 186(1), 1 (1999). F. Boccuzzi, J. Catal., 202(2), 256 (2001). H.S. Subbarao and E.C. Maiti, Solid State Ionics, 11, 317 (1984). C.G. Vayenas, S. Bebelis, C. Pliangos, S. Brosda and D. Tsiplakides, Electrochemical Activation of Catalysis: Promotion, Electrochemical Promotion, and Metal-Support Interactions. New York: Kluwer Academic/Plenum Publishers (2001). C. Vayenas, S. Brosda and C. Pliangos, J. Catal., 216(1-2), 487 (2003). P. Vernoux, M. Guth, and X. Li, Electrochem. Solid St., 12, E9 (2009). R.J. Isaifan, H.A.E. Dole, E. Obeid, L. Lizarraga, P. Vernoux and E.A. Baranova, Electrochem. Solid St., 15(3), E1 (2012). R.J. Isaifan, H.A.E. Dole, E. Obeid, L. Lizarraga, E.A. Baranova and P. Vernoux, ECS Trans., 35(28), 43 (2011). W. Heller and T.L. Pugh, J. Polym. Sci., 47(149), 203 (1960). M.Y. Han, C.H. Quek, W. Huang, C.H. Chew and L.M. Gan, Chem. Mater., 11(4), 1144 (1999). J. Wagner and J.M. Köhler, Nano Lett., 5(4), 685 (2005). H. Hirai, J. Macromol. Sci. A, 13, 633 (1979). T. Ashida, K. Miura, T. Nomoto, S. Yagi, H. Sumida, G. Kutluk, K. Soda, H. Namatame and M. Taniguchi, Surf. Sci., 601(18), 3898 (2007). Y. Leng, K. Sato, J.-G. Li, T. Ishigaki, M. Iijima, H. Kamiya, T. Yoshida, Powder Technol., 196(1), 80 (2009). L. Prati and G. Martra, Gold Bull., 32(3), 96 (1999). X. Zuo, H. Liu and M. Liu, Tetrahedron Lett., 39, 1941 (1998). J. Ayache, L. Beaurier, J. Boumendil, G. Ehret and D. Laub, “Chapter 5 Replica Technique,” in Sample Preparation Handbook for Transmission Electron Microscopy, 1st ed., Springer, p. 229-256 (2010). M. Boudart, Adv. Catal., 20, 153 (1969). M. Valden, X. Lai and D. Goodman, Science, 281(5383), 1647 (1998). C.T. Campbell, G. Ertl, H. Kuipers and J. Segner, J. Chem. Phys., 73(11), 5862 (1980). X. Su, P.S. Cremer, Y.R. Shen and G.A. Somorjai, J. Am. Chem. Soc., 119(17), 3994 (1997). K.R. McCrea, J.S. Parker and G.A. Somorjai, J. Phys. Chem. B, 106(42), 10854 (2002). S. Arrii, F. Morfin, A.J. Renouprez and J.L. Rousset, J. Am. Chem. Soc., 126(4), 1199 (2004). W. Li, M. Comotti and F. Schuth, J. Catal., 237(1), 190 (2006). E. Quinet, L. Piccolo, H. Daly, F.C. Meunier, F. Morfin, A. Valcarcel, F. Diehl, P. Avenier, V. Caps, J.-L. Rousset, Catal. Today, 138(1-2), 43 (2008). R.J.H. Grisel and B. E. Nieuwenhuys, J. Catal., 199(1), 48 (2001). M. Lamallem, H. El Ayadi, C. Gennequin, R. Cousin, S. Siffert, F. Aissi, A. Aboukais, Catal. Today, 137(2-4), 367 (2008).
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS274 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Catalytic CO Oxidation over Au Nanoparticles Supported on Yttria-Stabilized Zirconia Holly A.E. Dolea, Jong Min Kima, Leonardo Lizarragab, Philippe Vernouxb, Elena A. Baranovaa* a
Department of Chemical and Biological Engineering, University of Ottawa, 161 LouisPasteur St. Ottawa, ON, K1N 6N5 Canada b Université Lyon 1, CNRS, UMR 5256, IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon, 2 avenue Albert Einstein, F-69626 Villeurbanne, France *
email : [email protected]
Gold nanoparticles supported on yttria-stabilized zirconia (YSZ) were synthesized using a modified alcohol method which involves the stabilizing agent, polyvinylpyrrolidone (PVP). Two different average sizes (13.1 and 17.1 nm) of Au nanoparticles were synthesized. These catalysts were tested for gas phase CO oxidation. Since PVP is known to block the active Au sites, the catalysts were calcined at 300°C and 600°C in order to remove the PVP. Overall, higher catalytic activity was found for the smaller Au nanoparticles. It was also found that calcination is required in order to achieve activity, even though the particle size increases with this treatment. The effect of calcination temperature did not prove to be significant. It is demonstrated that O2- ionically conductive YSZ is a promising catalyst support that can finely disperse and stabilize Au nanoparticles for CO oxidation. Introduction For many years, gold has been disregarded as an active catalyst in its bulk state due to its chemically inert property. It wasn’t until Haruta (1-3) discovered that when gold particles with diameters smaller than 10 nm are supported on metal oxides, they tend to be very catalytic active for many reactions. However, this discovery came with the challenge of achieving small Au nanoparticles as Haruta stated, due to quantum size-effects, the smaller the Au nanoparticle, the lower the boiling point which results in significant agglomeration during calcination (1). Recently, research has been focused on determining the effect of size and pretreatment conditions (i.e., calcination temperature). It has been found that Au nanoparticles less than 5 nm show a significant increase in catalytic activity (4, 5). Others have also shown that calcination causes agglomeration and that as the temperature is increased the gold phase changes from gold oxide to metallic gold which shows to influence the catalytic activity (6, 7). The effect of the support material has been shown to have less of an influence on the activity (5) which could allow for flexibility depending on the desired application. The Au nanoparticles used in this study were supported on yttria-stabilized zirconia (YSZ), a metal oxide support. YSZ is known to be an O2-conductor due to the presence of oxygen vacancies inside its crystallographic structure. This oxide has been extensively studied as a material for oxygen sensors and fuel cells (8). Recently, YSZ has received attention as a catalytic support especially in studies of Electrochemical Promotion of Catalysis (EPOC); also called the NEMCA (Non Faradaic Electrochemical Modification of the Catalytic Activity) effect (9). Vayenas et al. (9) were the first to show that the migration of ionic species from a solid
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS265 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
electrolyte to the catalyst surface induced by electrical polarizations can improve catalytic performances. Vayenas et al. (10) have described EPOC as an electrically controlled strong metal support interaction (SMSI). Recently it was shown that YSZ is an attractive catalyst support for nanoparticle stabilization and activation for the reaction of CO oxidation over Pt/YSZ nanoparticles (11-13). It was found that O2- ionic species from YSZ can significantly promote the catalytic activity of Pt nanoparticles due to the thermally induced O2- backspillover from YSZ over Pt catalyst in agreement with the electrochemical promotion mechanism. It was also found that observed effect becomes more pronounced with decreasing the particle size (12). In this work, gold nanoparticles supported on YSZ (Au/YSZ) are investigated for the first time in the reaction of CO oxidation. To this end, Au nanoparticles with two different average sizes (13.1 and 17.1 nm) were synthesized using a modified alcohol method. This method was modified to achieve polyvinylpyrrolidone- (PVP) stabilized Au nanoparticles supported on YSZ. The variation in nanoparticle size was achieved by adjusting the ratio of Au to PVP in the synthesis solution (14-16). The particle sizes of these catalysts were studied using x-ray diffraction (XRD). Surface properties were also characterized using transmission electron microscopy (TEM). An effective calcination temperature was also determined for the removal of PVP using a thermogravimetric analysis (TGA). The PVP-stabilized Au/YSZ nanoparticles were tested using the model reaction of CO oxidation. The effect of the Au particle size and calcination temperature on the catalytic activity was investigated. Experimental Synthesis of Au Nanoparticles Gold nanoparticles (Au), deposited on YSZ powder (Au/YSZ) were synthesized using a similar method to the colloidal methods using alcohol as a reducing agent (17-19); however, the procedure was modified to achieve supported Au/YSZ. Gold (III) chloride (Aldrich, 99.999%) precursor salt was dissolved in a mixture of nanopure water (18 MΩ cm) and ethanol (Fischer, 99%) with a ratio of water to ethanol of 1:1, in the presence of PVP (Sigma, 10,000 average molecular weight) and YSZ powder (TOSOH, 8% YSZ, with average crystallite size of 25 nm, 0.6 µm particle size and bulk density of 1.3 g·cm-3). The molar ratio of PVP to gold was 3:1 (Au1) and 10:1 (Au2) for the samples tested. The desired gold loading was 1 wt. % on YSZ. The obtained mixture was stirred for 30 minutes at room temperature then refluxed at 100°C for 2 h, followed by extensive washing with nanopure (18 MΩ·cm) water and drying in the air at 70°C. The initial color of the cloudy mixture appears yellow. After about 30 minutes, the colour of the mixture turns pink or purple, depending on the concentration of PVP. Upon drying, the synthesized 1wt. % Au nanocatalysts on YSZ powder have a pink to purple color. PVP is used as a polymer protective agent in order to achieve small nanoparticles by stabilizing the Au ions during the synthesis (14-16). It is believed that the Au nanoparticles are stabilized through the reduction of van der Waals interaction between colliding colloidal particles (14). Before testing the performance of the catalysts, the PVP must be removed as it known to inhibit the catalytic activity (20, 21). This was done by calcining the Au/YSZ catalyst at 300 or 600°C for 2 h in air.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS266 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
Transmission Electron Microscopy Transmission electron microscopy (TEM) images were taken for the Au/YSZ catalysts tested (JEOL 2010 LaB6) in order to obtain a higher resolution so that the morphology and size of Au particles may be better investigated. An extraction replica technique was used for sample preparation (22). The catalyst was dispersed in ethanol, deposited on a mica film and covered with a carbon layer. The YSZ support and film of mica were then dissolved in hydrogen fluoride solution for 24 hours resulting in Au particles being fixed on the carbon film. These particles were directly observed by TEM. The use of Image J software allowed for the determination of Au particle size distribution. X-ray Diffraction X-ray diffraction (XRD) patterns of the Au/YSZ nanoparticles were collected using RigakuUltima IV diffractometer using Cu Ksource. The experiments were run in the focused beam geometry with a divergence slit of 2/3 degree, a scan speed of 0.17 degree·min-1 and a scan step of 0.06 degree for all experiments. The diffractograms were collected between 30 and 55°2. The average particle diameter was determined using the Scherrer equation and the full widths at half maximum (FWHM) of the (111) plane of gold at 38° 2. Thermogravimetric Analysis Thermogravimetric analysis (TGA) (2960 SDT V3.0F) was carried out to determine an optimum temperature range at which the synthesized Au/YSZ samples are to be calcined in order to completely remove the PVP stabilizing agent. The temperature range should be high enough to decompose the carbons and other volatile compounds in the PVP polymer, but also, the temperature should be low enough that sintering and formation of agglomerates of Au nanoparticles are minimized. In order to determine the proper calcination temperature, TGA is run to establish the temperature range at which carbon decomposes. The sample is heated at a rate of 1.2°C min-1 from room temperature to 580°C. Catalytic Activity Measurements The catalytic activity measurements of the Au/YSZ catalysts were carried out at atmospheric pressure in a continuous flow quartz reactor. The reaction gases were CO (Linde, 1.02% CO in He) combined with pure O2 (Linde, pure O2), and pure He (Linde, pure He) as a carrier gas. The gas compositions were controlled by mass flow controllers (Brooks, 5850 TR Series and MKS electronic mass flow controllers) and were the following: 760-780 ppm CO with 4.3-4.5% O2 and He balance for CO oxidation. The overall flow rate was held constant at 4.3 L·h-1, resulting in a space velocity of approximately 19 000 h-1. The product gases for the oxidation of CO were analyzed with an on-line micro-chromatograph (μGC-R3000 SRA instruments) which contained two separate columns; one to separate O, N2 and CO at a temperature of 90°C and the other to separate CO2 at a temperature of 100°C. CO2 concentration in the product gas of the reactor was also monitored by an infrared (IR) analyzer (Horiba, VA3000). A K-type thermocouple was attached to the reactor at the catalyst bed in order to monitor the temperature during the heat-up period. The entire reactor was placed in a furnace, which was attached to a temperature controller. The sample was heated from 25 to 250°C at a rate of
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS267 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
3°C/min. For each catalyst, three heating cycles were performed to observe the stability of the catalyst. Results and Discussion Characterization of Au Catalysts Figure 1 shows TEM images of Au2 which was untreated and calcined at 600°C. This image shows that the Au nanoparticles are uniformly distributed over the YSZ support. From the corresponding histograms, an average particle diameter was found to be 9.1 and 11.7 nm, respectively. These results show that calcination does have an effect on the particle size.
(a)
Percentage of Particles (%)
18 15 12 9 6 3 0 3
4
5
6
7
8
9 10 11 12 13 14
Particle Diameter (nm)
(b)
Percentage of Particles (%)
18 15 12 9 6 3 0 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Particle Diameter (nm)
Figure 1. TEM images (left) and corresponding histograms (right) of Au2 (a) untreated; and (b) calcined at 600°C.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS268 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
The XRD spectrum for each Au/YSZ catalysts is shown in Figure 2. The crystallite size for each sample was determined from the Au (111) peak using the Scherrer equation. These values are summarized in Table 1. It is evident that the size of the crystallite increased (approximately double) with calcination. Similar trends have been found in past studies (6, 7) where it is shown that calcination causes the gold nanoparticles to agglomerate as the melting temperature of the gold decreases with decreasing particle size.
Relative Intensity
4 3
(a) (b) (c) (d)
2 1 0 37 38 39 40 41 2[theta] (deg)
Figure 2. XRD spectra of Au (111) peak of Au/YSZ catalysts (a) Au2 calcined at 600°C; (b) Au1 calcined at 600°C; (c) Au2 untreated; and (d) Au1 untreated. Dispersion calculations were performed for both Au/YSZ catalysts using the average particle diameter from the XRD measurements and the following equation (23), dispersion (%) (MWAu 600) ( Au davg aAu Na )
[1]
where MWAu is the molecular weight of Au (196.97 g·mol-1), ρAu is the density of Au (19.32 g·cm-3), davg is the average particle size, aAu is the surface area of one Au atom (2.29 x 10-19 m2·atom-1), and Na is Avogadro’s number (6.022 x 1023 atom·mol-1). It is evident that since the crystallite size changes between untreated and calcined catalysts (approximately doubled), the dispersion also changes. However, these differences in dispersion do not appear to influence the catalytic activity as Au2 has lower dispersion but higher activity than Au1 (Figure 3). Table 1 summarizes these dispersion values as well as the particle sizes for each catalyst.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS269 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
1st Cycle 2nd Cycle 3rd Cycle Blank Experiment
CO conversion (%)
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 3. Catalytic performance for CO oxidation of Au2, calcined at 600°C. TABLE I. Summary of Au/YSZ Catalysts Prepared and Corresponding Characterization Data. CO Oxidation Avg. Particle Turnover Dispersion Activation Diameter Frequency, TOF Catalyst Pre-treatment (%) Energy, (s-1) (x10-3) davg (nm)* Ea (kJ·mol-1) 100°C 200°C Au1 untreated 8.4 7.1 0 0 77.0 Au1 calcined at 300°C 56.1 Au1 calcined at 600°C 17.1 3.4 0 78.7 63.6 Au2 untreated 6.3 5.3 0 0.25 93.2 Au2 calcined at 600°C 13.1 2.6 8.45 181.74 24.1 *from XRD measurements The result from the TGA of Au1 is shown in Figure 4. The first mass loss of 11.9 % is related to water evaporation and the second mass loss of 1.4 % is related to polymer decomposition. There is a slight loss of mass just below 600°C, likely being a partial removal of oxygen from YSZ. In order to evaluate the effect of all these losses detected from TGA, 300°C (removal of PVP only) and 600°C (removal of all species) selected to be the suitable temperature for calcination.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS270 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
19.5 19.0
Mass (mg)
18.5 11.9% 18.0 17.5 17.0 1.4%
16.5 0
100 200 300 400 500 600 o Temperature ( C)
Figure 4. Thermogravimetric analysis of Au1. Catalytic Performance for CO Oxidation To study the catalytic performance of the Au/YSZ catalysts, the oxidation of CO was used as a model reaction. This reaction has been widely studied due to its environmental and industrial significance (24-27). For each catalyst, the catalytic oxidation of CO was repeated for three successive heating ramps between 25 and 250°C. Figure 3 shows the catalytic activity of Au2 calcined at 600°C. The results showed that the catalyst exhibits stable performance after the first cycle where the activity is observed from around 65°C. It is observed that the first cycle shows lower catalytic activity compared to the successive cycles. Similar multi-cycle performance has been observed by other research groups (28-30); however, the opposite performance has also been observed (31, 32). This observation was explained by the dependence of the pretreatment conditions. In a previous study, it was determined that when catalysts were pre-oxidized, the catalytic activity of the first heating cycle was lower (31). It was explained that when the catalyst is exposed to a reducing species such as CO, the catalyst is activated which is shown by the higher catalytic activity in the successive runs. Another explanation could be the result of residual PVP even after calcination. It is possible that during the first cycle, there is PVP inhibiting the reaction, and it is only after the first cycle that the remaining PVP is removed. Effect of Particle Size Figure 5 shows the effect of Au particle size on the catalytic performance for CO oxidation. Two different ratios of PVP:Au were used to control the size of the particles which has also been shown in previous studies (16, 21). It was explained that the PVP can slow the diffusion of the gold ions within the solution, preventing them from coagulating; therefore, more PVP creates less diffusion. However, it has been shown that there is a limit to the amount of PVP that can influence the particle size (16). From this study, it is clear that the smaller gold
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS271 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
nanoparticles (Au2) synthesized using a ratio of 10:1 (PVP:Au) have the highest catalytic activity.
CO Conversion (%)
17.1 nm 13.1 nm Blank Experiment
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 5. Comparison of catalytic performance for CO oxidation of Au/YSZ catalysts, calcined at 600°C with two different average particle size. Effect of Calcination Temperature Since PVP is known to inhibit catalytic reactions (20, 21), the prepared Au/YSZ catalysts were pretreated by calcining at elevated temperatures to remove this stabilizing polymer. However, calcination can also have a negative effect on the performance of the catalysts. It has been previously shown that higher calcinations temperatures cause agglomeration of the nanoparticles (6, 7). Since smaller nanoparticles are desired for greater activity, the amount of PVP and calcination temperature must be optimized to achieve the best performance. Figure 6 shows a comparison of Au1 that was untreated, calcined at 300°C and calcined at 600°C with respect to a blank experiment (i.e. reactor without any catalyst). From the figure, it is evident that there is a significant effect on the catalytic activity between the untreated and calcined samples; however, the difference between calcinations temperatures does not appear to have such a significant effect. These results show that PVP does indeed inhibit the reaction. Since the PVP is used to stabilize the Au nanoparticles, it is said to be surrounding the nanoparticles. This in turn prevents the surface of the Au nanoparticles to be exposed resulting in lower catalytic activity. Therefore, it is important to sufficiently remove the PVP without causing significant agglomeration of the nanoparticles.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS272 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
(a) (b) (c) (d)
CO conversion (%)
100 80 60 40 20 0 80
120
160
200
240
280
Temperature (°C) Figure 6. Effect of calcination temperature on the catalytic activity for CO oxidation of Au1 (a) untreated; (b) calcined at 300°C; (c) calcined at 600°C; and compared to a (d) blank experiment. Conclusion It is demonstrated that YSZ is a promising catalyst support for Au nanoparticles for the reaction of CO oxidation. Au nano-catalysts of two average sizes were synthesized using a modified alcohol method with a PVP stabilizer. The size of the particles depended on the amount of PVP used – more PVP made smaller particles. The catalytic performance of the untreated and calcined catalysts was evaluated for the model reaction of CO oxidation. It was found that calcination had an effect by increasing the particle size by approximately double; however, this treatment was necessary to remove the PVP stabilizer which is known to inhibit the reaction. This was evident by the better performance of the calcined catalyst compared to the untreated catalyst. The most important factor proved to be the size of the particle, where the smaller nanoparticle (Au2) showed the highest catalytic activity for CO oxidation. This catalyst showed catalytic activity beginning around 90°C. In order to improve this catalytic activity, the size of the Au nanoparticles will need to be decreased in further studies. Acknowledgements The authors would like to thank the Natural Science and Engineering Research Council (NSERC) for financial support, the Centre for Catalysis Research and Innovation (University of Ottawa) for XRD and TGA analysis, and IRCELYON for microscopy services.
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS273 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Transactions, 45 (1) 265-274 (2012)
References [1] [2] [3] [4] [5] [6] [7] [8] [9]
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
[23] [24] [25] [26] [27] [28] [29] [30] [31] [32]
H. Haruta, Surf. Sci., 6(2), 102 (2002). M. Haruta, Chem. Rec., 3(2), 75 (2003). M. Haruta, ChemInform., 35(44), 27 (2004). J.D. Grunwaldt, C. Kiener, C. Wögerbauer and A. Baiker, J. Catal., 181(2), 223 (1999). N. Lopez, J. Catal., 223(1), 232 (2004). E.D. Park and J.S. Lee, J. Catal., 186(1), 1 (1999). F. Boccuzzi, J. Catal., 202(2), 256 (2001). H.S. Subbarao and E.C. Maiti, Solid State Ionics, 11, 317 (1984). C.G. Vayenas, S. Bebelis, C. Pliangos, S. Brosda and D. Tsiplakides, Electrochemical Activation of Catalysis: Promotion, Electrochemical Promotion, and Metal-Support Interactions. New York: Kluwer Academic/Plenum Publishers (2001). C. Vayenas, S. Brosda and C. Pliangos, J. Catal., 216(1-2), 487 (2003). P. Vernoux, M. Guth, and X. Li, Electrochem. Solid St., 12, E9 (2009). R.J. Isaifan, H.A.E. Dole, E. Obeid, L. Lizarraga, P. Vernoux and E.A. Baranova, Electrochem. Solid St., 15(3), E1 (2012). R.J. Isaifan, H.A.E. Dole, E. Obeid, L. Lizarraga, E.A. Baranova and P. Vernoux, ECS Trans., 35(28), 43 (2011). W. Heller and T.L. Pugh, J. Polym. Sci., 47(149), 203 (1960). M.Y. Han, C.H. Quek, W. Huang, C.H. Chew and L.M. Gan, Chem. Mater., 11(4), 1144 (1999). J. Wagner and J.M. Köhler, Nano Lett., 5(4), 685 (2005). H. Hirai, J. Macromol. Sci. A, 13, 633 (1979). T. Ashida, K. Miura, T. Nomoto, S. Yagi, H. Sumida, G. Kutluk, K. Soda, H. Namatame and M. Taniguchi, Surf. Sci., 601(18), 3898 (2007). Y. Leng, K. Sato, J.-G. Li, T. Ishigaki, M. Iijima, H. Kamiya, T. Yoshida, Powder Technol., 196(1), 80 (2009). L. Prati and G. Martra, Gold Bull., 32(3), 96 (1999). X. Zuo, H. Liu and M. Liu, Tetrahedron Lett., 39, 1941 (1998). J. Ayache, L. Beaurier, J. Boumendil, G. Ehret and D. Laub, “Chapter 5 Replica Technique,” in Sample Preparation Handbook for Transmission Electron Microscopy, 1st ed., Springer, p. 229-256 (2010). M. Boudart, Adv. Catal., 20, 153 (1969). M. Valden, X. Lai and D. Goodman, Science, 281(5383), 1647 (1998). C.T. Campbell, G. Ertl, H. Kuipers and J. Segner, J. Chem. Phys., 73(11), 5862 (1980). X. Su, P.S. Cremer, Y.R. Shen and G.A. Somorjai, J. Am. Chem. Soc., 119(17), 3994 (1997). K.R. McCrea, J.S. Parker and G.A. Somorjai, J. Phys. Chem. B, 106(42), 10854 (2002). S. Arrii, F. Morfin, A.J. Renouprez and J.L. Rousset, J. Am. Chem. Soc., 126(4), 1199 (2004). W. Li, M. Comotti and F. Schuth, J. Catal., 237(1), 190 (2006). E. Quinet, L. Piccolo, H. Daly, F.C. Meunier, F. Morfin, A. Valcarcel, F. Diehl, P. Avenier, V. Caps, J.-L. Rousset, Catal. Today, 138(1-2), 43 (2008). R.J.H. Grisel and B. E. Nieuwenhuys, J. Catal., 199(1), 48 (2001). M. Lamallem, H. El Ayadi, C. Gennequin, R. Cousin, S. Siffert, F. Aissi, A. Aboukais, Catal. Today, 137(2-4), 367 (2008).
Downloaded on 2014-06-16 to IP 137.122.64.44 address. Redistribution subject to ECS274 terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
